2026 heat-pump brand share in Europe: who wins on R290 vs R32?

Europe’s 2026 heat-pump market in one snapshot

The latest EPREL snapshot on Househeating Pulse contains 60,989 listed heat-pump models from 777 manufacturers (market_index_snapshot / Househeating Pulse · Market Index v1, computed from EPREL Public API). That is the right starting point for a Europe-wide comparison of visible model portfolios, especially in the live EPREL catalog and the broader manufacturer index. It is not, however, a sales ledger. EPREL records declared products, not shipped units or installed base.

At the whole-market level, the average listed heat pump in the snapshot carries a SCOP of 4.55, an average declared output of 9.3 kW, and average outdoor sound power of 61.3 dB (market_index_snapshot / Househeating Pulse · Market Index v1, computed from EPREL Public API). By type, the registry is led by air-water with 30,452 models, followed by air-air with 21,065, heat-pump water heaters with 9,228, ground-water with 213, and water-water with 31 (market_index_snapshot / Househeating Pulse · Market Index v1, computed from EPREL Public API). Readers comparing subsegments can cross-check those slices in the market snapshot and filtered air-to-water catalog.

The refrigerant picture is highly uneven. EPREL declares 13,935 models using R32 and 537 using R290 (market_index_snapshot / Househeating Pulse · Market Index v1, computed from EPREL Public API). Together, R32 and R290 account for 14,472 listings, or 23.73% of all 60,989 listed models (computed from market_index_snapshot / Househeating Pulse · Market Index v1, computed from EPREL Public API). Within that pair, R32 alone represents 22.85% of all listings, while R290 represents 0.88% (computed from market_index_snapshot / Househeating Pulse · Market Index v1, computed from EPREL Public API).

That low R290 count matters. The platform’s broader natural-refrigerant share is just 3.27% of listed models (market_index_snapshot / Househeating Pulse · Market Index v1, computed from EPREL Public API), and the declared-code table shows why careful interpretation is needed: EPREL also contains minor spelling variants such as R290A with 2 listings and R290a with 1 listing, while R410A still appears on 1,896 listings and R410 on 10 (refrigerant_universe / IPCC AR6 GWP table; EU Reg. 2024/573 phase-out schedule; EPREL declared codes). For regulatory context, the EU F-gas Regulation 2024/573 reference in the corpus marks R32 with a phase-out date of 2027-01-01 and R410A with 2025-01-01, while R290 has no listed phase-out date (refrigerant_universe / IPCC AR6 GWP table; EU Reg. 2024/573 phase-out schedule; EPREL declared codes). Readers wanting the property table rather than the model counts should use the site’s refrigerants reference.

Who dominates EPREL by brand count

The overall leaderboard is concentrated, and very top-heavy. The top 5 brands hold 51.93% of all listed models, the top 10 hold 60.88%, and the top 20 hold 67.26% (computed from brand_share / EPREL Public API · brand-share aggregation). That means roughly two-thirds of the visible EPREL market is controlled by just 20 manufacturers, despite the registry containing 777 manufacturers in total (market_index_snapshot / Househeating Pulse · Market Index v1, computed from EPREL Public API).

Top 10 manufacturers by EPREL model count

The gap at the top is stark. Daikin alone accounts for 24.05% of all EPREL-listed heat-pump models, compared with 9.14% for Mitsubishi Electric Europe and 8.54% for Johnson Controls Hitachi (brand_share / EPREL Public API · brand-share aggregation). That overall dominance is easy to verify in the leaderboards hub or directly in the filtered Daikin catalog.

R290 vs R32: the refrigerant split that reshuffles leadership

The full-market ranking does not tell the whole story, because the refrigerant split is lopsided. As noted above, EPREL currently shows 13,935 R32 models and only 537 R290 models (refrigerant_universe / IPCC AR6 GWP table; EU Reg. 2024/573 phase-out schedule; EPREL declared codes). So any brand comparison inside R32 listings and R290 listings is comparing very different pools of declared products.

The corpus supports a strong editorial point here but not a complete refrigerant-segment leaderboard. It clearly shows that refrigerant strategy matters, because the reference table identifies R290 as a natural hydrocarbon with GWP 0 and R32 as an HFC with GWP 771 (refrigerant_universe / IPCC AR6 GWP table; EU Reg. 2024/573 phase-out schedule; EPREL declared codes). It also shows that the overall market is still numerically dominated by R32 declarations rather than R290 declarations.

What the corpus does not include is a manufacturer-by-refrigerant breakdown. There is no probe here giving brand counts within the R290-only or R32-only slices, no top-10 R290 brand ranking, no top-10 R32 brand ranking, and no per-brand refrigerant mix for the overall top 10. The registry snapshot therefore cannot support numeric claims such as “Brand X leads R290 with Y models” or “Brand Z gains N percentage points in R32” from the material provided. Those are precisely the kinds of claims that would require an explicit brand-by-refrigerant aggregation from EPREL.

The same limitation applies to the requested question about how much leading brands lose or gain when the market is split by refrigerant. The corpus establishes the overall leaderboard and the refrigerant totals, but not the bridge between them.

Which brands win in each segment

On the evidence available, only one segment winner can be identified with confidence: Daikin Europe N.V. leads the overall EPREL market with 14,668 listed models and 24.05% share (brand_share / EPREL Public API · brand-share aggregation). Its average SCOP is 4.44 (brand_share / EPREL Public API · brand-share aggregation).

For the refrigerant-specific segments, the registry extract supplied here does not record brand winners. The two top_models probes for R290 and R32 return no data rows (top_models / EPREL Public API via Househeating Pulse catalog), and the brand_detail probe failed entirely (brand_detail / (probe failed — data unavailable)). So the corpus cannot answer, with numbers, which manufacturer leads R290 or which leads R32.

The same applies to refrigerant-specific efficiency by brand. The corpus does provide average SCOP for the leading brands overall — for example Bosch Thermotechnik GmbH at 4.69, Ariston SpA at 4.66, and Mitsubishi Electric Europe B.V. at 4.51 (brand_share / EPREL Public API · brand-share aggregation). It also shows that among the top 20 overall, LG Electronics Deutschland GmbH posts 4.93, STIEBEL ELTRON GmbH & CO. KG 4.84, and Viessmann Climate Solutions SE 4.61 (brand_share / EPREL Public API · brand-share aggregation). But there is no per-brand SCOP split by refrigerant, so the registry extract cannot show whether high-volume brands are more efficient in their R290 portfolios than in their R32 portfolios.

Readers interested in raw efficiency rankings should use the site’s top SCOP leaderboard, plus the more specific air-to-water SCOP ranking and ground-source SCOP ranking. Those pages answer a different question: best listed efficiency by model, not biggest manufacturer portfolio by refrigerant.

What concentration tells us about market structure

Even without the missing cross-tab, EPREL already shows a concentrated market structure. The largest brand alone holds 24.05% of all listed models, while the next two brands add 9.14% and 8.54% respectively (brand_share / EPREL Public API · brand-share aggregation). The top three therefore control 41.73% of all listings (computed from brand_share / EPREL Public API · brand-share aggregation). Add Bosch and Ariston, and the top five reach 51.93% (computed from brand_share / EPREL Public API · brand-share aggregation).

That concentration sits alongside a long tail of smaller registrants. After the top 10, the next 10 brands together add only 6.38 percentage points, taking the cumulative share from 60.88% for the top 10 to 67.26% for the top 20 (computed from brand_share / EPREL Public API · brand-share aggregation). In plain terms: EPREL is broad enough to contain hundreds of manufacturers, but the declared model universe is still dominated by a relatively small group.

Efficiency also varies across the big names. Among the overall top 10, average SCOP ranges from 4.18 for Johnson Controls Hitachi to 4.69 for Bosch Thermotechnik (brand_share / EPREL Public API · brand-share aggregation). That spread is meaningful, but it should not be over-read as a winner table for end users. SCOP averages are portfolio averages across mixed product types and capacities, not like-for-like comparisons. Buyers should pair brand-level data with model-level checks in the full catalog, and installers can sharpen shortlist decisions with the sizing calculator and payback calculator.

What EPREL can and cannot say about real sales

EPREL is strong on portfolio visibility. It can show how many models are listed, how concentrated the manufacturer field is, which refrigerants appear in declarations, and what average attributes look like across the database (market_index_snapshot / Househeating Pulse · Market Index v1, computed from EPREL Public API). It is therefore useful for comparing breadth of offer, especially when readers move between the methodology page, the glossary, and the live catalog filters.

EPREL is weak on commercial market share. The registry does not record units sold, installed base, replacement cycles, channel mix, or country-by-country shipments in the corpus provided. So “brand share” in this article means share of listed models in EPREL, not share of European sales. Likewise, the declared refrigerant counts are counts of registered models, not evidence that 13,935 R32 units or 537 R290 units were sold. The corpus itself frames those numbers as declared usage counts (refrigerant_universe / IPCC AR6 GWP table; EU Reg. 2024/573 phase-out schedule; EPREL declared codes).

That distinction matters even more for policy interpretation. The refrigerant reference table can tell readers that R290 has GWP 0, R32 has GWP 771, and that R32 has a listed phase-out date of 2027-01-01 under the cited schedule (refrigerant_universe / IPCC AR6 GWP table; EU Reg. 2024/573 phase-out schedule; EPREL declared codes). It cannot tell readers how quickly any brand will shift its shipped volume from one refrigerant to another. For that, shipment data or audited market reports would be needed.

For readers comparing Europe by geography rather than by manufacturer, the next stop is the 32-country comparison dashboard and the country index. But for a registry-based view of product breadth in 2026, the main takeaway is narrower and more defensible: the EPREL universe is concentrated overall, and the refrigerant question is central enough that any serious brand comparison needs a dedicated R290-versus-R32 cut rather than a single all-market leaderboard.

Sources

• Househeating Pulse · Market Index v1, computed from EPREL Public API — snapshot date 2026-05-20
• EPREL Public API · brand-share aggregation — snapshot date 2026-05-20
• IPCC AR6 GWP table; EU Reg. 2024/573 phase-out schedule; EPREL declared codes — snapshot date 2026-05-20
• EPREL Public API via Househeating Pulse catalog — snapshot date 2026-05-20
• (probe failed — data unavailable) — no valid snapshot date supplied

Continue reading

• Heat pump SCOP explained — How to read seasonal efficiency figures without confusing test conditions and real-world use.
• R290 vs R32 heat pumps — A practical primer on the trade-offs between propane and difluoromethane systems.
• How to compare heat pumps in EPREL — A method for separating registry noise from meaningful model differences.
• Air-to-water heat pump sizing basics — A concise sizing framework before narrowing down brands or refrigerants.